Abstract Age estimation is a crucial component in the identification of individuals. The London Atlas is a radiographic method for dental age estimation, and it is non-invasive, easy to apply, and widely utilized. Previous studies confirmed the high accuracy of the London Atlas in Northern and Central Thai groups. However, this study aimed to evaluate its accuracy in a southern Thai population since genetic and environmental differences occur among populations. A total of 360 digital panoramic radiographs from Thai children aged between 7 and 15 years (180 males and 180 females) were analyzed. The sample was divided into nine age groups. Each group included 40 radiographs. Dental age was estimated by comparing the radiographs with the London Atlas. Differences between chronological and dental ages were assessed using the Wilcoxon signed rank test. Intra- and inter-observer reliability scores were 0.839 and 0.816, respectively. The mean dental age estimated using the London Atlas (11.57 years) slightly exceeded the actual mean chronological age (11.52 years). However, the difference was not statistically significant in either sex (P > 0.05). The London Atlas method showed a minor overestimation of 0.01 years in males and 0.09 years in females. In conclusion, the London Atlas method demonstrated a high level of accuracy and is appropriate for dental age estimation in the southern Thai population.

Keywords: Dental age estimation, London Atlas, Panoramic radiograph, Thai population, Tooth development

Citation:  Duangto, P., Phandusa, S., Pratchyadechakul, N., Wankam, P., Charoemratrote, C., and Theerasopon, P. 2026. Dental age estimation using the London Atlas in a southern Thai population. Natural and Life Sciences Communications. 25(2): e2026037.

Graphical Abstract:

INTRODUCTION

Age estimation can be conducted using various methods that include skeletal age assessment, evaluation of secondary sexual characteristics, and analysis of tooth development (Nayyar et al., 2016). Skeletal development and secondary sexual characteristics are significantly influenced by factors such as nutrition, hormones, and the environment that lead to considerable variation (Prentice, 2001; Lashin et al., 2019; Manotas et al., 2022; Lashin et al., 2024). In contrast, tooth development is primarily determined by genetic factors and is less affected by external influences compared to the skeletal system.

Teeth are among the most durable structures in the human body, which exhibit resistance to environmental changes, chemical degradation, and high temperatures (Chai et al., 2009; George et al., 2017). Therefore, teeth serve as a reliable tool for age estimation in both living and deceased persons (Theerasopon et al., 2024; Lashin et al., 2025).

Dental age estimation can be performed primarily by evaluating tooth development, changes in fully formed teeth, and alterations in the surrounding tissues (Swami et al., 1992). These methods can be broadly categorized into three main groups.

First, the visual method involves direct observation of the teeth. It is simple, non-invasive, and easy to apply. However, this approach provides limited information and cannot reveal internal details such as root formation or stages of tooth development. As a result, it lacks precision and is not highly accurate when used as the single method for dental age estimation.

Second, the biochemical method relies on measuring chemical changes in dental tissues, such as the racemization of aspartic acid in dentin. This method offers high accuracy but is invasive, which makes it unsuitable in living individuals. It is primarily applicable to deceased people and therefore cannot be used in all cases.

Third, the radiographic method is the most widely used approach for dental age estimation (Schmeling et al., 2004; Theerasopon et al., 2023). This method offers several advantages as it is simple, noninvasive, does not require tooth extraction, allows for repeatable assessments, and is applicable to both living and deceased individuals. Additionally, radiographic imaging provides detailed information, which includes the complete structure of the teeth and surrounding tissues. Therefore, the radiographic method is considered highly reliable, accurate, and widely suitable for use in dental age estimation (Theerasopon et al., 2023).

Dental age estimation using the London Atlas represents a major advancement in forensic odontology by providing a clear and standardized method based on dental development. The London Atlas was developed by AlQahtani, Hector, and Liversidge in 2010 at Queen Mary University of London (AlQahtani et al., 2010). The London Atlas was based on a large, contemporary dataset, which makes it more accurate and relevant than earlier standards such as those by Schour and Massler or Ubelaker (AlQahtani et al., 2014).

The London Atlas method uses panoramic radiographs to assess the development of both deciduous and permanent teeth, as well as alveolar eruptions, which allows age estimation from 28 weeks in utero to 23.99 years. It includes 31 age categories with tooth-specific illustrations and median developmental stages that enable direct visual comparison between radiographs and the atlas images. The London Atlas is freely available in multiple languages and uses color-coded diagrams for enamel, dentine, and pulp to improve clarity and ease of use. Its practicality, non-invasiveness, cost-effectiveness, and reproducibility make it a valuable tool in forensic, archaeological, and anthropological contexts where precise age estimation is crucial (AlQahtani et al., 2014).

Due to genetic and environmental differences among various population groups, the accuracy and applicability of dental age estimation methods can vary across ethnicities. Previous studies have demonstrated that the London Atlas provides high accuracy and generally satisfactory results in several populations worldwide, which have included the Northern and Central Thai populations (Namwong and Mânica, 2020; Duangto et al., 2023). These findings suggest that population-specific validation is essential to ensure the reliability of age estimation methods. Therefore, the aim of this study was to evaluate the accuracy of dental age estimation using the London Atlas in a Southern Thai population.

MATERIALS AND METHODS

This retrospective study was approved by the Human Research Ethics Committee of the Faculty of Dentistry, Prince of Songkla University, Songkhla, Thailand (EC 6803-012), and informed consent was waived. Digital panoramic radiographs (Figure 1) were acquired using the GXDP-700 PANOREX + Cone Beam machine (Gendex Dental Systems, Hatfield, PA, USA). The radiographs were collected from the archives of the Faculty of Dentistry, Prince of Songkla University, which spanned the years 2015 to 2024.

Figure 1. Digital panoramic radiograph of a sample.

The sample size was calculated using G*Power software version 3.1.9.7 by a matched pair t-test with an effect size of 0.2, a study power of 95%, and the significance level was set at 5%. The calculated sample size was at least 327 samples. Digital panoramic radiographs of 360 Thai children (180 males and 180 females) aged between 7 and 15 years were randomly selected and divided into nine age groups with 40 radiographs per group (Table 1).

Table 1. Distribution of the samples by age group and sex.

The inclusion criteria required Thai nationality of all subjects, complete demographic data, which included names, sex, and dates of birth, high-quality digital radiographs with radiographic dates, and clear visibility of dental structures. All information was recorded and kept confidential.

The exclusion criteria included non-Thai individuals, unclear radiographs, systemic diseases, pathologies of the face and jaw, patients who had received or were currently undergoing orthodontic treatment, patients with severe malocclusion, dental anomalies such as hypodontia and hyperdontia, and other pathological conditions affecting tooth development. Chronological age was calculated based on the difference between the date of birth and the radiograph date and expressed in years with two decimal places.

The study specifically focused on Thai children aged 7 to 15 years old, which is a critical developmental period characterized by mixed dentition and the maturation of permanent teeth. This age range was selected to effectively evaluate the transition from deciduous to permanent dentition and the completion of root formation.

Dental age was assessed by directly comparing the radiographs with the London Atlas (AlQahtani et al., 2010). Each sample of dental age was compared with the chronological age. The difference was calculated by subtracting chronological age from the dental age, where positive values indicated overestimation and negative values indicated underestimation. The mean error (ME) and mean absolute error (MAE) were recorded.

To evaluate intra- and inter-observer reliability, 50 radiographs were randomly selected after the initial assessment. The first researcher (S.P.) reassessed these radiographs to determine intra-observer reliability, while a second researcher (P.W.) independently assessed them to determine inter-observer reliability. Cohen’s kappa test was used for the statistical analysis.

Descriptive statistics, including the mean and absolute differences between dental and chronological ages, were calculated by another researcher (N.P.). The Wilcoxon signed rank test was used to compare dental and chronological ages across all age groups. A significance level of 0.05 was applied for hypothesis testing. All statistical analyses were performed using the SPSS software package (SPSS for Windows, version 27, Chicago, IL, USA).

RESULTS

Cohen’s kappa values were 0.839 for intra-observer reliability and 0.816 for inter-observer reliability, which indicated almost perfect reproducibility (Table 2).

Table 2. Values of intra-observer reliability and inter-observer reliability using Cohen’s kappa test.

The mean chronological age (11.52 years) and the mean dental estimated age (11.57 years) for the combined samples showed no statistically significant difference (P = 0.095) (Table 3). When analyzed separately by sex, a significant difference was not observed. Males had a mean chronological age of 11.52 years and a dental age of 11.53 years with a difference of 0.01 years (P = 0.495) (Table 4), while females had a mean chronological age of 11.51 years and a dental age of 11.61 years with a difference of 0.09 years (P = 0.098) (Table 5).

Overall, the London Atlas tended to overestimate age in males (except in the 13-, 14-, and 15-year-old groups) and to underestimate age in females (except in the 8-, 9-, 10-, and 14-year-old groups). However, comparisons within most individual age groups revealed no statistically significant differences (P > 0.05).

Table 3 compares dental age estimated by the London Atlas with chronological age for both sexes combined. In the 7-, 8-, 9-, and 10-year-old groups, the mean estimated dental age was slightly overestimated, while in all other age groups, dental age was underestimated compared to chronological age.

The overall mean absolute error (MAE) between dental age and chronological age was 0.39 years when both sexes were combined (Table 3). When analyzed separately, the MAE was 0.38 years for males and 0.39 years for females (Tables 4 and 5). These findings demonstrated that the London Atlas method provides reliable and consistent age estimation results for both males and females.

Table 3. Comparisons between the London Atlas dental age (DA) and chronological age (CA) for both sexes.

Note: n: number of samples, ME: mean error, MAE: Mean absolute error, SD: Standard deviation, *Statistically significant difference using Wilcoxon signed rank test (P < 0.05).

Table 4. Comparisons between the London Atlas dental age (DA) and chronological age (CA) for males.

Note: n: number of samples, ME: mean error, MAE: Mean absolute error, SD: Standard deviation, *Statistically significant difference using Wilcoxon signed rank test (P < 0.05).

Table 5. Comparisons between the London Atlas dental age (DA) and chronological age (CA) for females.

Note: n: number of samples, ME: mean error, MAE: Mean absolute error, SD: Standard deviation, *Statistically significant difference using Wilcoxon signed rank test (P < 0.05).

DISCUSSION

Age estimation plays a crucial role in forensic science, which is an interdisciplinary field that has been integrated into the legal system. It is particularly significant in identifying deceased individuals or victims of mass disasters, as it aids in narrowing down the search for missing persons (Lewis and Senn, 2010; Gelbrich et al., 2020).

Chronological age is central to both civil and criminal proceedings, as many legal rights and responsibilities are age-dependent. This is especially important in cases involving juvenile offenders or victims (Ghonem et al., 2024). Therefore, accurate age determination is essential in upholding fairness and justice in legal matters. In situations where the date of birth is unknown, biological indicators can be evaluated to estimate an individual’s age.

Estimating age through dental development is considered more reliable than skeletal features or secondary sexual characteristics, as the latter are more susceptible to environmental influences such as nutrition and hormonal levels. In contrast, tooth development is largely governed by genetic factors (Prentice, 2001; Manotas et al., 2022). Moreover, teeth are highly durable and resistant to environmental conditions such as chemical degradation and high temperatures (Chai et al., 2009; George et al., 2017). As a result, teeth serve as a valuable tool for age estimation in both living individuals and the deceased.

There are various methods for dental age estimation. Radiographic techniques are used most often, and they are reliable. These methods are non-invasive and offer comprehensive insights by displaying the full structure of the teeth and surrounding tissues.

Radiographic methods are broadly categorized into atlas-based methods and scoring methods. Scoring methods, such as those by Demirjian or Willems, involve assigning stages to multiple teeth and converting them into a score or age through mathematical formulas. While precise, these can be time-consuming and may require population-specific adaptations. In contrast, the atlas methods utilize a visual comparison of the entire radiographic image, among which the London Atlas method stands out due to its simplicity and wide applicability across diverse populations.

Several atlas methods are available that include the Schour and Massler method, Ubelaker’s charts, and the London Atlas. The London Atlas was selected for this study because it was developed using a larger and more contemporary dataset that makes it more accurate and relevant for modern populations. Age estimation using the London Atlas involves comparisons of panoramic radiographs with the corresponding London Atlas chart. This approach offers a simple, rapid, cost-effective, and consistent method for age estimation.

However, atlas methods have some limitations that rely on the subjective interpretation of the observer and represent age as a median developmental category rather than a continuous variable, which may overlook subtle individual variations. However, the London Atlas method demonstrated good reproducibility in this study in both intra-observer and inter-observer reliability levels.

The accuracy of this method has been tested in various ethnic groups around the world that included populations from Western, Asian, and African regions (Jacometti et al., 2023; Chua, et al., 2025). Previous studies have consistently reported a low mean error of less than one year to support its reliability, which agreed with the results of our study.

The findings of our study demonstrated that the London Atlas offers highly accurate dental age estimations in relation to chronological age within this population. The absence of statistically significant differences between chronological and dental ages, both in the overall sample (P = 0.095) and when analyzed separately by sex (males: P = 0.495; females: P = 0.098), indicates that the method reliably estimates age. The mean differences observed (i.e. 0.01 years for males and 0.09 years for females) are minimal and further support the method’s accuracy.

Although there was a slight tendency to overestimate age in males and underestimate age in females, these patterns were not statistically significant in most individual age groups. The overall mean absolute error values of 0.38 years for males, 0.39 years for females, and 0.39 years for both sexes combined demonstrated the high accuracy and consistency of the London Atlas. These results are consistent with previous research that showed the London Atlas delivers precise age estimations across different populations.

Our results demonstrated acceptable accuracy for age estimation using the London Atlas in the combined-sex Thai samples with a mean error of 0.05 years. This value is notably lower than the results reported in previous studies, such as Namwong and Manica’s (-0.50 year) and Duangto et al. (0.07 year). Our findings confirmed that the London Atlas is a highly effective method for dental age estimation in the samples examined in this study.

Although there are differences in mean error values, the results of this study are consistent with previous research confirming the overall reliability of the London Atlas as an effective method. Additionally, like several other studies, this research observed slight variations in mean error values between males and females. This pattern has also been reported in studies involving Turkish (Koç et al., 2021), New Zealand (Baylis and Based, 2017), and Hispanic (McCloe et al., 2018) populations.

While this study confirms the high accuracy of the London Atlas in a southern Thai population, several limitations should be considered. First, the evaluation relies on a visual comparison method, which introduces a degree of subjectivity despite the high intra- and inter-observer reliability scores achieved in this research.

Second, the London Atlas uses median developmental stages for age categories, which may not fully capture the continuous nature of biological growth or every individual variation within a specific age group. Finally, the sample was restricted to children and adolescents aged 7 to 15 years; therefore, the findings may not be generalizable to infants or older young adults.

CONCLUSION

This study demonstrated the accuracy of the London Atlas of human tooth development and eruption for estimating age using digital panoramic radiographs of Thai children and adolescents from 7 to 15 years of age. A strong correlation was demonstrated between dental age and chronological age in both males and females. Importantly, the combined-sex analysis revealed no statistically significant difference between the mean chronological age and the mean dental age. These findings support the high accuracy and suitability of the London Atlas for dental age estimation in the southern Thai population.

ACKNOWLEDGEMENTS

The authors gratefully acknowledge the Dental Hospital, Faculty of Dentistry, Prince of Songkla University, Songkhla, Thailand and School of Dentistry, University of Phayao, Phayao, Thailand for their support and facilitation of this study.  

AUTHOR CONTRIBUTIONS

Phuwadon Duangto: Conceptualization (Lead), Methodology (Lead), Formal Analysis (Lead), Validation (Lead), Resource (Equal), Writing – Original Draft (Lead), Writing – Review & Editing (Lead), Investigation (Equal), Supervision (Lead), Project Administration (Equal); Solya Phandusa: Data Curation (Equal), Formal Analysis (Equal), Writing – Original Draft (Equal), Writing – Review & Editing (Equal), Investigation (Lead); Nannapat Pratchyadechakul: Data Curation (Equal), Formal Analysis (Equal), Writing – Original Draft (Equal), Writing – Review & Editing (Equal), Investigation (Lead); Pannita Wankam: Data Curation (Equal), Formal Analysis (Equal), Writing – Original Draft (Equal), Writing – Review & Editing (Equal), Investigation (Lead); Chairat Charoemratrote: Methodology (Supporting), Formal Analysis (Supporting), Validation (Equal), Resource (Lead), Data Curation (Lead), Writing – Review & Editing (Equal), Investigation (Supportive), Supervision (Equal), Project Administration (Supportive); Pornpat Theerasopon: Conceptualization (Lead), Methodology (Lead), Formal Analysis (Equal), Validation (Lead), Resource (Equal), Writing – Original Draft (Lead), Writing – Review & Editing (Lead), Investigation (Equal), Supervision (Lead), Project Administration (Lead).

CONFLICT OF INTEREST

The authors declare that they hold no competing interests.

REFERENCES

AlQahtani, S.J., Hector, M.P., and Liversidge, H.M. 2010. The London atlas of human tooth development and eruption. American Journal of Physical Anthropology. 142(3): 481-490. https://doi.org/10.1002/ajpa.21258

AlQahtani, S.J., Hector, M.P., and Liversidge, H.M. 2014. Accuracy of dental age estimation charts: Schour and Massler, Ubelaker and the London Atlas. American Journal of Physical Anthropology. 154(1): 70-78. https://doi.org/10.1002/ajpa.22473

Baylis, S. and Bassed, R. 2017. Precision and accuracy of commonly used dental age estimation charts for the New Zealand population. Forensic Science International. 277: 223-228. https://doi.org/10.1016/j.forsciint.2017.06.011

Chai, H., Lee, J.J., Constantino, P.J., Lucas, P.W., and Lawn, B.R. 2009. Remarkable resilience of teeth. Proceedings of the National Academy of Sciences of the United States of America. 106(18): 7289-7293. https://doi.org/10.1073/pnas.0902466106

Chua, A., Halim, C., Pham, E., Samuel, S., Zafar, S., and AlQahtani, S. 2025. Reliability of London Atlas for dental age estimation in an Australian cohort. International Journal of Legal Medicine. 139(6): 2913-2920. https://doi.org/10.1007/s00414-025-03559-2

Duangto, P., Janhom, A., and Iamaroon, A. 2023. Age estimation using the London Atlas in a Thai population. Australian Journal of Forensic Sciences. 55(6): 700-707. https://doi.org/10.1080/00450618.2022.2040588

Gelbrich, B., Carl, C., and Gelbrich, G. 2020. Comparison of three methods to estimate dental age in children. Clinical Oral Investigations. 24(7): 2469-2475. https://doi.org/10.1007/s00784-019-03109-2

George, R., Tan, W.J., Shih Yi, A.L., and Donald, P.M. 2017. The effects of temperature on extracted teeth of different age groups: A pilot study. Journal of Forensic Dental Sciences. 9(3): 165-174.

Ghonem, M.M., Eldeeb, B.S., Elaggan, A.M., and Lashin, H.I. 2024. Age estimation based on sternebral fusion in a sample of Egyptian population using three-dimensional computed tomography scanning. Forensic Imaging. 39: 200613. https://doi.org/10.1016/j.fri.2024.200613

Jacometti, V., Sato, C.M., Meireles, D.A., and Silva, R.H.A.D. 2023. Age estimation using London Atlas methodology: A systematic review and meta-analysis. Forensic Science International. 342: 111532. https://doi.org/10.1016/j.forsciint.2022.111532

Koç, A., Özlek, E., and Öner Talmaç, A.G. 2021. Accuracy of the London Atlas, Willems, and Nolla methods for dental age estimation: A cross-sectional study on eastern Turkish children. Clinical Oral Investigations. 25(8): 4833-4840. https://doi.org/10.1007/s00784-021-03788-w

Lashin, H.I., Eldeeb, B.S., and Ghonem, M.M. 2019. Sex identification from Foramen Magnum using computed tomography scanning in a sample of Egyptian population. Journal of Forensic Radiology and Imaging. 19: 100341. https://doi.org/10.1016/j.jofri.2019.100341

Lashin, H.I., Zaki, E.A., Hafez, A.S., Reda, A.M. and Kabbash, A.M.M. 2024. Development of a novel humerus-clavicle model for sex determination using multi-detector computed tomography scanning in a sample of Egyptian population. Forensic Imaging. 38: 200601. https://doi.org/10.1016/j.fri.2024.200601

Lashin, H.I., Sharif, A.F., Ghaly, M.S., El-Desouky, S.S. and Elhawary, A.E. 2025. Bridging gaps in age estimation: A cross-sectional comparative study of skeletal maturation using Fishman method and dental development using Nolla method among Egyptians. International Journal of Legal Medicine, 139(2): 695-714. https://doi.org/10.1007/s00414-024-03394-x

Lewis, J.M. and Senn, D.R. 2010. Dental age estimation utilizing third molar development: A review of principles, methods, and population studies used in the United States. Forensic Science International. 201(1-3): 79-83. https://doi.org/10.1016/j.forsciint.2010.04.042

Manotas, M.C., González, D.M., Céspedes, C., Forero, C., and Rojas Moreno, A.P. 2022. Genetic and epigenetic control of puberty. Sexual Development. 16(1): 1-10. https://doi.org/10.1159/000519039

McCloe, D., Marion, I., da Fonseca, M.A., Colvard, M., and AlQahtani, S. 2018. Age estimation of Hispanic children using the London Atlas. Forensic Science International. 288: 332.e1-332.e6. https://doi.org/10.1016/j.forsciint.2018.04.013

Namwong, W. and Mânica, S. 2020. Testing the London Atlas for age estimation in Thai population. Acta Odontologica Scandinavica. 78(3): 161-164. https://doi.org/10.1080/00016357.2019.1667526 

Nayyar, A.S., Babu, B.A., Krishnaveni, B., Devi, M.V., and Gayatri, H.C. 2016. Age estimation: Current state and research challenges. Journal of Medical Sciences. 36(6): 209. https://doi.org/10.4103/1011-4564.196348

Prentice, A. 2001. The relative contribution of diet and genotype to bone development. The Proceedings of the Nutrition Society. 60(1): 45-52. https://doi.org/10.1079/PNS200072

Swami, D., Mishra, V.K., Bahal, L., and Rao, C.M. 1992. Age estimation from eruption of temporary teeth in Himachal Pradesh. International Journal of Forensic Medicine and Toxicological Sciences. 9: 3-7. 

Schmeling, A., Olze, A., Reisinger, W., and Geserick, G. 2004. Forensic age diagnostics of living people undergoing criminal proceedings. Forensic Science International. 144(2-3): 243-245. https://doi.org/10.1016/j.forsciint.2004.04.059

Theerasopon, P., Charoemratrote, C., and Duangto, P. 2023. Age estimation using the Demirjian and Willems methods in a southern Thai population. Natural and Life Sciences Communications. 22(4): e2023058. https://doi.org/10.12982/NLSC.2023.058

Theerasopon, P., Woratanarat, W., Saiwong, N., Netprakon, N., Charoemratrote, C., and Duangto, P. 2024. Accuracy of age estimation method using dental radiographs of permanent mandibular second molars in a southern Thailand population. Natural and Life Sciences Communications. 23(3): e2024030. https://doi.org/10.12982/NLSC.2024.030

OPEN access freely available online

Natural and Life Sciences Communications

Chiang Mai University, Thailand. https://cmuj.cmu.ac.th